[09:28] <evand> http://pastebin.ubuntu.com/107262/ - this is what is really triggering the encryption option crash underneath the blacklisting issue.  Curiously, if I stick a sudo chroot in front of that it works just fine.
[09:28] <evand> ^ kirkland
=== davmor2 is now known as davmor2-lunch
=== davmor2-lunch is now known as davmor2
[14:33] <davmor2> evand: xivulon: I'm testing hardy.2 and I'm still getting invalid cd could it be that I am running the iso on a dvd re-writable?
[14:34] <davmor2> on umenu and wubi is still trying to download
[14:36] <evand> is D:/disk/.info (or whatever the CD drive is) present?
[14:36] <evand> err
[14:36] <evand> .disk/info
[14:42] <davmor2> If I run it in cmd it says no recognised app or batch program however I can view it in notepad
[14:43] <CIA-59> oem-config: cjwatson * r578 trunk/ (5 files in 2 dirs):
[14:43] <CIA-59> oem-config: Simplify page/step handling; the list of pages is now in BaseFrontend,
[14:43] <CIA-59> oem-config: and step indices are calculated directly from that. (This won't quite
[14:43] <CIA-59> oem-config: work for Ubiquity as-is, because it can't yet handle multi-page steps
[14:43] <CIA-59> oem-config: such as Ubiquity's partitioner.)
[14:44] <davmor2> evand: ^
[14:45] <xivulon> davmor2: type "%temp%" in explorer and grab the wubi log in there, it should explain why it is downloading (or put it in the pastebin and post the link)
[14:49] <davmor2> xivulon: hang on I just tried it again umenu still no joy but wubi is at least reading from the cd now
[14:52] <davmor2> evand: .disk/info reads ubuntu 8.04.2 " Hardy Heron" - Release amd64 (20090120) if that helps
[14:52] <CIA-59> oem-config: cjwatson * r579 trunk/ (63 files in 3 dirs):
[14:52] <CIA-59> oem-config: Reword templates, package descriptions, and the oem-config(8) manual
[14:52] <CIA-59> oem-config: page to make it clearer that the first stage of oem-config is for use by
[14:52] <CIA-59> oem-config: OEMs, not end users (LP: #48144). Clarify the term "OEM" in a comment
[14:52] <CIA-59> oem-config: for translators.
[14:53] <CIA-59> oem-config: cjwatson * r580 trunk/debian/control: adjust oem-config-udeb description further to make it a less tempting selection
[14:53] <davmor2> just to confirm wubi is reading from the cd quite happily it's just umenu that throws up an invalid cd notice
[15:02] <evand> hrm, very odd
[15:04] <evand> I'll pull down the .2 CD and see if I can get to the bottom of it.
[15:06] <evand> (might be a little while;  virgin cable is nonsense, in my opinion)
[15:06] <davmor2> evand: you in the uk then?
[15:07] <evand> ja
[15:07] <evand> moved over this past weekend
[15:07] <davmor2> Ah okay not me going mad then :)
[15:07] <evand> get to move again this weekend (just up the road a bit)
[15:07] <evand> heh
[15:08] <davmor2> evand: if it helps I can drop it on my server :)
[15:08] <evand> thanks, but I seem to be getting the same speed just about everywhere.  I wonder if I am being rate limited.
[15:10] <davmor2> evand: you'll notice a drop when the schools are out
[15:10] <evand> yikes
[15:10] <evand> what time is that usually?
[15:10] <davmor2> about now
[15:10] <evand> ah
[15:10] <evand> fun
[15:12] <davmor2> virgin is on demand and it limited.  If you download more than so much between 11am and 9pm I think you drop to 5meg.  But if loads of people in the same area start using iplayer for instance then it cripples the speed across the board
[15:14] <evand> yeah, we debated between Virgin and Be Unlimited, but went with the former on account of her already being tied to a virgin account for TV and because of the eventual 50 Mb rollout.  The excessive rate limiting has me a bit worried though, as I suspect I can hit that with a daily DVD download.
[15:15] <cjwatson> heh, discovering the joys of UK net access?
[15:16] <evand> oh yeah
[15:16] <davmor2> evand: http://allyours.virginmedia.com/html/internet/traffic.html
[15:16] <evand> they had rolled out fiber to the home before I left the states.
[15:17] <evand> davmor2: indeed, I've read through that a few times :)
[15:17] <davmor2> evand: I go for a cronjob each morning to grab the latest cd's and dvd's before the blight sets in :)
[15:18] <cjwatson> a local mirror might be the answer
[15:18]  * evand files a expense report for a direct link to the datacenter
[15:18] <cjwatson> (it's what I do; my ADSL down is pathetic)
[15:18] <evand> of cdimage?  Wouldn't James come after me?
[15:18] <cjwatson> of the archive
[15:18] <cjwatson> then jigdo
[15:18] <evand> ah, indeed
[15:18] <cjwatson> it's not perfect but it does save me a certain amount of aggro
[15:19] <davmor2> you don't get penalised that way unless there are several re-spins in the day :)
[15:19] <cjwatson> you still need to be able to download the 1.1GB template for DVDs or whatever it is, though
[15:19] <cjwatson> I use http://people.ubuntu.com/~cjwatson/tmp/jaunty-daily-jigdo which is hacky as well but more or less works
[15:19] <cjwatson> *as hell
[15:20] <evand> hrm, I'll look into it once I have the desktop and external drives unpacked.  This poor little laptop certainly doesn't have the space for an archive mirror
[15:20] <davmor2> cjwatson: run it over night the first time then cronjob it daily from then
[15:20] <cjwatson> er, and rsyncs over ssh *cough*
[15:20] <evand> hahaha
[15:20] <evand> nice; thanks
[15:24] <davmor2> evand: just to be certain as soon as I finish this other install I'm running I'll run a test against vista too (I'm expecting it to be the same)
[15:24] <evand> ok
[15:37] <CIA-59> ubiquity: evand * r2983 ubiquity/ (3 files in 2 dirs):
[15:37] <CIA-59> ubiquity: Full screen the window when in ONLY_UBIQUITY mode. The extra screen
[15:37] <CIA-59> ubiquity: real estate makes the timezone map, among other widgets, easier to
[15:37] <CIA-59> ubiquity: use.
[15:40] <davmor2> evand:  just to throw a spanner in the works too I just tried m-a the partition knew that xp was on the machine but m-a didn't
[15:56] <lfaraone> evand: although first I need to figure out why the code doesn't run, is the latest bzr version of usb-creator working on intrepid, or do I have to be on jaunty?
[15:56] <lfaraone> Traceback (most recent call last):  File "./bin/usb-creator", line 75, in <module> f = GtkFrontend(options.iso,options.persistent)
[15:56] <lfaraone> TypeError: __init__() takes exactly 1 argument (3 given)
[15:58] <evand> is this from bzr branch lp:usb-creator?
[15:59] <lfaraone> evand: yes. All I've done is bzr branch and push to my copy.
[15:59] <evand> interesting.  Does usbcreator/gtk_frontend.py have something other than:
[16:00] <lfaraone> evand: odd, I just rebranched in ~/usb-creator and it works.
[16:00] <evand> def __init__(self,iso=None,persistent=True):
[16:00] <evand> ah, ok
[16:00] <lfaraone> evand: I was in a directory inside a few symlinks.
[16:00] <evand> ok
[16:03]  * lfaraone hacks.
[16:13] <davmor2> evand: m-a says there were no users or operating systems suitable for importing from. :(
[16:13] <evand> hrm
[16:13] <evand> amd64?
[16:13] <davmor2> again I'll try with vista
[16:13] <davmor2> amd64 yes
[16:14] <evand> vista wont work
[16:14] <davmor2> oh no it was intrepid were it started too wasn't it :(
[16:14] <evand> there's an outstanding bug where m-a doesn't work properly on amd64, though I have to look into it.
[16:14]  * evand remembers the pet-bug scheme
[16:15] <davmor2> evand: okay will try 32bit.  It used to work and I think it did in jaunty testing could be wrong though I lost track
[16:15] <evand> ok; thanks
[16:31] <davmor2> evand: amd64 invalid cd on vista too
[16:32] <evand> noted, still pulling down the cd
[16:36] <davmor2> evand: m-a is working on 32bit
[16:37] <evand> ah, then it's probably bug 218673
[16:37] <ubottu> Launchpad bug 218673 in migration-assistant "64bit Migration-Assistant can't see XP" [High,Confirmed] https://launchpad.net/bugs/218673
[16:38] <davmor2> evand: you'd think I'd remember my own bug wouldn't you D'oh
[16:39] <evand> :)
[16:41] <davmor2> you get so used to it working in intrepid + that you forget it didn't once upon a time :)
=== robbiew1 is now known as robbiew
[16:43] <tjaalton> cjwatson: have you had time to look at the patch in bug 315363?
[16:43] <ubottu> Launchpad bug 315363 in pkgsel "add an option to install "Recommended" packages too" [Wishlist,Triaged] https://launchpad.net/bugs/315363
[16:50] <xivulon> davmor2: try to runa umenu --debug
[16:51] <xivulon> should give you lots of messages on what is happening and some explanation
[16:51] <davmor2> xivulon: cool thanks for the info
[17:08] <davmor2> xivulon: evand: umenu is looking for Ubuntu .....  In .disk/info it's listed as ubuntu.....  will that make a difference?
[17:43] <cjwatson> tjaalton: I think it should be pkgsel/install-recommends rather than pkgsel/install_recommends, since pkgsel's other templates use hyphens, and you don't generally need to terminate shell commands with semicolons ('inst_rec="true";'), but otherwise it's fine - want to go ahead and commit that to bzr and upload it?
[17:44] <cjwatson> (there's another change in bzr too)
[18:03] <evand> davmor2: I can confirm the umenu bug.
[18:03]  * evand digs
[18:03] <davmor2> evand: I haven't bugged it yet was sure what to put it under just made note of it
[18:30] <lfaraone> I know this isn't a devlopment question, but is there a command line parameter to use a SD card as a home directory when booting from the livecd? (I'm trying to test out suspend/resume in jaunty without instaling it)
[18:33] <tjaalton> cjwatson: ah ok, I'll fix that and commit&upload tomorrow, thanks :)
[18:37] <evand> lfaraone: add persistent to the kernel command line (f6 at the boot menu), and make sure there is a ext2/ext3 loopback fs named home-rw in the root of the SD card.
[18:38] <evand> The filesystem on the SD card either needs to be vfat, or labeled "home-rw"
[18:39] <evand> (the code for this is casper.  bzr branch lp:casper, look at scripts/casper and scripts/casper-helpers)
[18:45] <lfaraone> evand: Thanks.
[18:45] <evand> anytime
[18:46] <lfaraone> evand: will jaunty casper support ext4? :)
[18:48] <cjwatson> ... it hadn't even occurred to me that casper would need to be fixed for ext4, but of course it does
[18:48] <cjwatson> evand: go ahead and make the obvious fixes if you feel like it :)
[18:50] <evand> will do
[18:52] <lfaraone> Excellent.
[18:57] <evand> cjwatson: http://pastebin.ubuntu.com/107418/ - thumbs up?
[18:57] <cjwatson> evand: aye
[18:58] <CIA-59> casper: evand * r569 casper/ (4 files in 3 dirs): Add support for ext4.
[19:02] <cjwatson> StevenK: what should armel daily builds be set up for - ubuntu-mid, ubuntu-umpc, or both?
[19:04] <lfaraone> evand: ok, so there is a brand new partition label'd "home-rw" on /dev/mmcblk0p1 and I booted with 'persistent' _before_ the "--" in the kernel parameters, should it have been after?
[19:04] <lfaraone> evand: (because right now I don't think it's really being persistent, /dev/mmcblk0p1 is only mounted in /media
[19:06] <evand> it wont matter if it's before or after, but anything you put after the -- will be included in the kernel command line on the installed system (should you choose to install)
[19:06] <evand> does it appear in /sys/block?
[19:06] <lfaraone> evand: Yes, /sys/block/mmcblk0
[19:09] <evand> hrm, you'll want to do some debugging using set -x in the casper scripts, and break=top in the kernel command line
[19:09] <evand> note that your only editor in that initial initramfs shell is sed
[19:10] <lfaraone> evand: god help me. :_
[19:11] <lfaraone> evand: would being a SDHC card complicate things?
[19:11] <evand> heh, I'd give you a hand, but I don't have a free machine to work with at the moment (kvm's usb support is almost absent in Ubuntu)
[19:11] <evand> shouldn't, AIUI
[19:11] <lfaraone> evand: ok, and I don't need a loopmounted file if the partition itself is ext(2,3)? (trying to see if this is user error)
[19:13] <evand> correct
[19:13] <evand> assuming that the partition is labeled home-rw
[19:14] <lfaraone> evand: it is. /me tries with a USB stick.
[19:14] <cjwatson> StevenK: and what does the form factor want to be - are we talking about a vfat-wrapped livefs, or straight d-i (netboot only, in which case why does antimony need to be involved?), or what?
[19:14] <evand> lfaraone: does vol_id say it is?
[19:14] <evand> ah, ok
[19:15] <evand> dinner, back in a bit
[19:15] <lfaraone> evand: No idea, but I used gparted to set the label and it's mounted to "/media/home-rw" automagically (both the sd and the usb)
[19:16] <evand> lfaraone: just for sanity's sake, check to make sure vol_id reports it as such, as that's what the actual test uses
[19:16] <evand> ah actually, if it's ending up as /media/home-rw, it should be fine
[19:16] <evand> as I'm fairy certain that code uses vol_id
[19:19] <lfaraone> Doesn't work with USB key either. Now I'm really wondering if it's User Error :)
[19:22] <lfaraone> Yep, blkid confirms: /dev/sdb1: LABEL="home-rw" UUID="601309f7-9016-49d4-99dc-5f72931f9d16" TYPE="ext3"
[19:47] <davmor2> evand: any joy with the hardy.2 umenu issue?
[20:20] <evand> lfaraone: you might want to give a casper-rw file (make the entire filesystem read/write) a try, as that's a often tested code path.
[20:20] <evand> davmor2: none yet, Windows is acting up and not giving me a network connection
[20:21] <davmor2> See and people wonder why we stick with linux :)
[20:23] <evand> heh
[20:32] <lool> cjwatson: I see d-i failed to build on armel due to size constraints of the nslu2 partitioning; is it ok to just change the partitioning and grant more space to the kernel and less to the initramfs?  It seems the initramfs uses half of its 6 MB and the kernel is more in the 2 MB ballpark than 1.3
[20:33] <lool> My understanding is that d-i creates a flashable image, and so I don't think changing the layout is an issue for installed devices, but perhaps there's a hardcoded partition layout on this flash which I don't know about
[20:35] <cjwatson> lool: yeah, I'd seen the build failure and was planning to look tomorrow :-) it sounds plausible but I have not investigated enough yet to be able to venture an opinion ...
[20:36] <lfaraone> evand: ok, I'll look into it.
[20:36] <lool> cjwatson: Right, I was hoping you had a clue about this particular hardware; I don't have it myself
[20:43] <evand> hrm, appears to be a kvm issue
[20:59] <superm1> cjwatson, i was looking at bug 290398, and I tried your comment but got a traceback about NoneType not having an attribute name (which makes sense).  isn't that if statement logic wrong then?  shouldn't it be "if cachedpkg is not None and not cached_pkg(cache, pkg)"?  or is there really a valid case where None could be in self.langpacks?
[20:59] <ubottu> Launchpad bug 290398 in ubiquity "Incomplete language support even with all languages preseeded" [Undecided,Incomplete] https://launchpad.net/bugs/290398
=== robbiew is now known as robbiew-afk
[21:41] <CIA-59> ubiquity: superm1 * r2984 ubiquity/ (debian/changelog scripts/install.py):
[21:41] <CIA-59> ubiquity: Adjust logic for detecting incomplete language support. Just
[21:41] <CIA-59> ubiquity: because a package doesn't "exist" doesn't mean that language support
[21:41] <CIA-59> ubiquity: isn't all there. Some packages don't yet provide language-support-XY
[21:41] <CIA-59> ubiquity: packages. (LP: #290398)
[21:41] <CIA-59> ubiquity: In this particular case, language-support-aa doesn't exist, but it is earlier
[21:41] <CIA-59> ubiquity: implicitly declared to exist when language-support-aa was added to the list.
[22:12] <CIA-59> wubi: Agostino Russo * r56 trunk/ (60 files in 13 dirs):
[22:12] <CIA-59> wubi: * Strip cpuid binary to reduce size
[22:12] <CIA-59> wubi: * Set the version and revision via the makefile
[22:12] <CIA-59> wubi: * Verify metalink md5 signature using openpgp and python-crypto
[22:12] <CIA-59> wubi: * Check md5 of blobs installed within wine during compilation
[22:12] <CIA-59> wubi: * Skip console logging when running as a packaged executable (pythonw)
[23:12] <StevenK> cjwatson: There's a spec about the armel builds -- they should all be vfat, with no kernel, and ubuntu should be the first thing to set them up for.
[23:12] <StevenK> cjwatson: The spec is https://wiki.ubuntu.com/Specs/ARMImageSelection
[23:37] <cjwatson> StevenK: aha, thanks
[23:38] <cjwatson> StevenK: rather than a --skip-kernel option, how about we just teach livecd.sh directly that armel doesn't have a kernel? that seems safer
[23:38] <StevenK> cjwatson: That's what I was thinking, but then debian-cd needs to be taught the same lesson
[23:39] <cjwatson> it needs to be taught something about armel anyway *shrug*
[23:39] <StevenK> Heh, fair enough